Placebo Effect and Expectations

My mindfulness and neurofeedback study focused on the placebo effect and the impact of expectations and rituals on wellness. The centerpiece of my research is a 2014 study by a group of researchers affiliated with Harvard Medical School and Beth Israel Deaconess Medical called “Altered Placebo and Drug Labeling Changes the Outcome of Episodic Migraine Attacks.” It was published in Science Translational Medicine (6:218ra5) in January 2014. Led by Slavenka Kam-Hansen, the research team planned to examine a series of questions related to the influence of context on the efficacy of the treatment for migraine, which, as a disorder with subjective, self-reported symptoms, is thought to be susceptible to placebo effects.

66 chronic migraine sufferers were observed over the course of seven acute migraine events each and asked to report pain on a scale of 1-10 at 0.5 and 2.5 hours after the onset of each event. The first migraine served as a baseline record; no treatment was provided. For each of the subsequent six migraines, received treatment at 0.5 hours after onset: a labeled envelope with an unlabeled pill inside. The envelope label would either read “placebo,” “Maxalt,” (a well known and effective migraine pain reliever), or it would be blank. The pill inside the envelope was either a genuine Maxalt, or it was a placebo. Each subject got every one of the six label-pill combinations ([placebo]-[Maxalt], [placebo-placebo], [blank]-[Maxalt], etc.) over the course of the six migraines, but they were delivered in random order. This design allowed the researchers to observe the effects of both treatment and placebo on migraines under varying expectations (on the part of the subject) that he or she was receiving active treatment.

The experiment turned out a number of very interesting findings. Here are some of the highlights:

First, as expected, more certainty in labeling (which translates to greater positive expectations for treatment) resulted in more reduction in pain — both for active treatment and placebo pills.

Second, amazingly, the efficacy of Maxalt mislabeled as placebo was not significantly better than placebo mislabeled as Maxalt. This suggests that information conveyed to a patient during a treatment can be as critical as the underlying medicine in determining outcomes.

Finally, and most surprisingly, open-label placebo ([placebo]-[placebo]) was much superior to the baseline. That is, when the subject took no treatment, the pain worsened, but when they took a placebo that was labeled as “placebo,” (a medicinally equivalent treatment) the pain improved. In fact, the open-label placebo improvement was 60% of the magnitude of the open-label Maxalt improvement, suggesting that the placebo effect could account for more than half of the efficacy of the migraine treatment!

The careful design of this study helps to illustrate that more focus should be given to managing patient expectations when attempting to treat subjective symptoms. Doctors have an ethical obligation not to mislead patients, but providing more positive information about treatments may be in the best interests of the patient if the information improves efficacy. Maybe there is an opportunity for the class to consider prototypes that would help doctors to deliver a more positive message when describing a treatment — perhaps by analyzing the doctor’s speech for positive or negative sentiment. Or maybe the design of pill bottles or other instruments involved in treatment can be re-thought to maximize the positive expectations that they create for patients interfacing with them.

Slavenka Kam-Hansen, Moshe Jakubowski, John M. Kelley, Irving Kirsch, David C. Hoaglin, Ted J. Kaptchuk, and Rami Burstein. “Altered Placebo and Drug Labeling Changes the Outcome of Episodic Migraine Attacks.” Sci Transl Med 8 January 2014 6:218ra5. [DOI:10.1126/scitranslmed.3006175]

Sciencemag.com: http://stm.sciencemag.org/content/6/218/218ra5.full

MIT Libraries: http://libproxy.mit.edu/login?url=http://search.ebscohost.com/login.aspx?direct=true&db=mdc&AN=24401940&site=eds-live